Objetivo: Verificar a acurácia do Test of Infant Motor Performance (TIMP) e da ultrassonografia de crânio (USC) no diagnóstico neurológico precoce após os 10 meses de idade corrigida em recém-nascidos pré-termo (RNPT) Metodologia: Amostra não aleatória constituída por 59 RNPT (idade gestacional ao nascimento 32 semanas ou peso ao nascimento 1500 gramas) seguidos em média até os 12 meses de idade corrigida. Os resultados da USC foram agrupados em 3 intervalos: 1) de 0 a 15 dias, 2) de 16 a 30 dias e 3) de 31 a 45 dias. Os achados da USC foram classificados em normal e anormal (anormalidades moderada e grave). O TIMP foi aplicado mensalmente, do primeiro retorno após a alta hospitalar até o 4° mês de idade corrigida. As avaliações foram agrupadas em 5 intervalos, correspondentes às avaliações antes do termo, 1°, 2°, 3° e 4° meses de idade corrigida. Os resultados do TIMP foram classificados em normal (média e média baixa) ou anormal (abaixo da média e muito abaixo da média). A avaliação neurológica foi realizada em média aos 12 meses de idade corrigida, e usada como padrão-ouro. Foram calculados os valores de sensibilidade, especificidade e valores preditivos positivos (VPP) e negativos (VPN) para o TIMP e para a USC neonatal. Resultados: A paralisia cerebral foi diagnosticada em 6 crianças. Observamos que a USC apresentou alta sensibilidade (> 70%) assim como altos VPN (>88%) em todos os intervalos. Para a USC, especificidade e VPP foram baixos em todos os intervalos. A sensibilidade do TIMP foi baixa, exceto para o intervalo 0, e os VPP foram baixos em todas as idades. A escala TIMP apresentou alta especificidade (75%, 85%) no 3° e 4° meses e altos VPN (> 77%) em todos os intervalos. Conclusão: Concluímos que os RNPT com pontuação normal no 3° e 4° meses do TIMP tem grandes chances de não desenvolver PC enquanto que RNPT com anormalidades graves e persistentes à USC tem maiores chances de um prognóstico neurológico anormal / Objective: Calculate the accuracy of the Test of Infant Motor Performance (TIMP) and the cranial ultrasonography (CUS) in the neurological outcome after 10 months of corrected age of preterm infants. Methods: Non-random sample of 59 preterm newborn infants (gestational age 32weeks or birth weight1500g) were followed up to a mean of 12 months corrected age. CUS results were grouped into 3 periods: 1) from 0 to 15 days; 2) from 16 to 30 days, and 3) from 31 to 45 days of life. CUS findings were rated into two groups: normal and abnormal (moderate and severe abnormalities). TIMP was applied monthly, from the first outpatient visit after hospital discharge until four months corrected age. The evaluations were grouped into five intervals, corresponding to the assessments performed before term age, 1st, 2nd, 3rd and 4th month of corrected age. TIMP results were ranked as normal (average, low average) or abnormal (below average and far below average). A full neurological examination was performed at a mean of 12 months of corrected age, and used as gold standard. The sensitivity, specificity, positive predictive (PPV) and negative predictive (NPV) values for TIMP and CUS were calculated. Results: Cerebral palsy was diagnosed in six infants. We observed that CUS had a high sensitivity (> 70%) in all intervals as well as high NPV (>88%). For CUS, specificity and PPV were low in all intervals. TIMP sensitivity was low, except for interval 0, and PPV were low at all ages. TIMP scale showed high specificity in the 3rd and 4th month (75%, 85%) and high NPV (> 77%) at all ages. Conclusions: We conclude that preterm infants with normal score at the 3rd and 4th months of TIMP are likely to develop normally while infants with severe and persistent abnormalities in the CUS examinations are more likely to have an abnormal neurological outcome

Reconnu comme l’une des principales causes d’invalidité, le traumatisme crânio-cérébral (TCC) est une condition chronique pouvant occasionner des déficits physiques, cognitifs, émotionnels et comportementaux. Ces déficits interfèrent avec la capacité de la personne à s’engager dans ses activités quotidiennes, incluant des activités complexes comme la préparation de repas. Afin de soutenir cette population lors de la préparation de repas, une technologie d’assistance à la cognition (TAC) nommée Cognitive Orthosis for coOKing (COOK) a été développée en partenariat avec une résidence alternative pour des personnes ayant des incapacités graves à la suite d’un TCC. Suivant l’installation de COOK, trois résidents ont été en mesure de préparer des repas de façon sécuritaire et autonome. Toutefois, rien n’est connu sur la façon avec laquelle COOK pourrait répondre aux besoins des personnes ayant eu un TCC et vivant dans la communauté, ni comment cette technologie pourrait offrir de l’assistance personnalisée, minimale et contextualisée pour soutenir cette population. Cette thèse vise donc à développer des connaissances pour soutenir le développement d’une version bonifiée de COOK qui correspondrait aux besoins variés de ces utilisateurs, en répondant à trois objectifs : 1) définir les besoins d’assistance verbale des personnes ayant eu un TCC modéré à grave afin d’améliorer les capacités de personnalisation de COOK; 2) évaluer son utilisabilité et expérience utilisateur en laboratoire et en milieu réel; et 3) identifier les facilitateurs et obstacles pouvant influencer l’implantation de COOK auprès de personnes ayant eu un TCC et vivant dans la communauté. Suivant une démarche de conception centrée sur l’utilisateur, quatre études ont été réalisées : (1) une étude mixte avec transformation des données et analyse d’enregistrements vidéo pour documenter l’assistance verbale offerte à 45 participants ayant un TCC lors de la préparation d’un repas; (2) une étude d’utilisabilité mixte pendant laquelle 10 participants ayant un TCC ont testé COOK en laboratoire; (3) une étude mixte à sujet unique pour évaluer l’utilisabilité de COOK au domicile d’une personne ayant eu un TCC grave; et (4) une étude qualitative descriptive impliquant des personnes ayant eu un TCC, des personnes proches aidantes et des intervenants, pour explorer leurs perceptions quant aux facilitateurs et obstacles à l’implantation de COOK dans leur milieu respectif. La première étude a permis d’identifier neuf types d’assistance cognitive et un type d’assistance motivationnelle, et de documenter les difficultés pour lesquelles les ergothérapeutes offraient de l’assistance. Des enjeux d’utilisabilité ont ensuite été documentés pendant les essais en laboratoire (étude 2) et en milieu réel (étude 3), particulièrement au niveau du système de sécurité autonome. Plusieurs besoins de bonifications technologiques ont aussi été identifiés par les participants (études 2, 3 et 4). Enfin, la quatrième étude a identifié divers facilitateurs (p.ex. les fonctionnalités disponibles dans COOK) et obstacles (p.ex. la logistique entourant l’installation, l’apprentissage et le soutien) à l’implantation de TAC comme COOK pour soutenir les personnes vivant dans la communauté avec un TCC grave. Ultimement, ces connaissances cliniques pourront soutenir le développement continu et l’implantation de TAC comme COOK pour qu’elles répondent aux besoins d’une plus grande population. / Defined as one of the main causes of disability, traumatic brain injury (TBI) is a chronic condition that can cause physical, cognitive, emotional, and behavioral deficits. These deficits may interfere with the person's ability to engage in daily activities, including complex activities such as meal preparation. To support this population during meal preparation, an assistive technology for cognition (ATC) called Cognitive Orthosis for coOKing (COOK) was recently developed in collaboration with an alternative residence for people with severe TBI. Following the installation of COOK, three residents were able to prepare meals safely and independently within their home. However, little is known about how COOK could meet the needs of people with TBI living in the community, or how it could provide personalized, minimal and contextualised assistance to support this population. This thesis aimed to develop a knowledge base to support the development of an improved version of COOK that would meet the varied needs of this new population, by meeting three objectives: 1) define the verbal assistance needs of people with moderate to severe TBI to enhance COOK's customization possibilities; 2) evaluate the usability and user experience of COOK in a laboratory setting and in a real-world environment; and 3) identify the facilitators and obstacles that may influence the implementation of COOK within the homes of people with TBI and in clinical settings. Integrated in a user-centered design process, four studies were conducted: (1) a study using conversion mixed design and video data analysis to document the verbal assistance provided to 45 participants with TBI during a meal preparation task; (2) a usability mixed-methods study during which 10 participants with TBI tested COOK in a laboratory setting; (3) a mixed-methods single-case study to assess the usability of COOK with a man with severe TBI living in the community; and (4) a qualitative descriptive study involving people with TBI, caregivers and healthcare professionals, to explore their perceptions of the facilitators and obstacles to the potential implementation of COOK in their respective settings. The first study identified nine types of cognitive assistance and one type of motivational assistance that were provided by occupational therapists to support people with TBI, and documented the difficulties for which these assistances were provided. Several usability issues were also documented when testing COOK in a laboratory setting (study 2) and in a real-world environment (study 3), particularly with the self-monitoring security system. Modification needs were also identified by the participants (studies 2, 3, and 4) to improve COOK. Finally, the fourth study highlighted many facilitators (e.g., the functionalities available in COOK) and obstacles (e.g., the logistics surrounding installation, learning and support of COOK) to consider before implementing ATCs like COOK. Ultimately, this knowledge base will support the continued development and the implementations of ATCs such as COOK to meets the needs of a larger population.